Answers are grounded in this paper and cite their sources.
High-resolution seafloor topography is essential for ensuring maritime navigation safety and providing data support for cutting-edge research in fields such as geodesy and geodynamics. Currently, marine gravity anomalies are the primary data source used in topography inversion. However, technical limitations in satellite altimetry result in insufficient accuracy of gravity models in reef regions; conventional inversion algorithms are mostly suitable for areas with gentle topography and perform inadequate in regions with significant undulations such as seamounts. To address these issues, this study employs the latest SWOT_03 gravity anomaly model and incorporates prior information from reef-shallow waters depth model and seamount morphology to develop a refined multisource data fusion algorithm for topography inversion that accounts for spatial heterogeneity. The algorithm is applied to South China Sea (SCS), a region characterized by widespread seamounts and reefs, and inversion experiments are conducted using both V32.1 gravity anomaly model and SWOT_03 gravity anomaly model. Results show that the inversion derived from the SWOT_03 model achieves a root mean square error (RMSE) error of 71.7 m when validated against ship soundings, compared to an RMSE of 99.1 m for a model constructed without integrating seamount and reef data. Moreover, the inversion results using SWOT_03 model yields a 14.6% improvement (71.7 m versus 84.0 m) compared to V32.1 model. Evaluated against the SRTM15+ and GEBCO_2024 topography models, the SWOT_03 enhanced inversion reduces RMSE by 51.4% (71.7 m versus 147.4 m) and 48% (71.7 m versus 136.8 m), respectively. These findings demonstrate that although refined marine gravity anomaly models contribute significantly to inversion accuracy, their performance remains limited in regions with distinctive topography features. Integrating multiple independent data sources to accurately characterize such specific topographies offers a promising pathway for further advancing topography models.
No takes yet. Share an insight, caveat, or question.
Xu et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: